旋转机械全息序列相似性匹配故障诊断方法

Holospectrum series similarity matching for rotating machinery fault diagnosis

针对全息诊断分辨率低影响旋转机械故障诊断质量和自动化水平的问题,将时间序列相似性匹配的基本概念和方法引入故障诊断应用中,结合全息诊断信息融合分析旋转机械振动全貌的思想,定义了全息序列及其相似性度量模型,用类时间轴上的多维序列表征转子系统振动全貌,进而利用采用近似三角不等式与B+树结合剪枝策略的全息序列相似性匹配算法实现故障诊断。实验结果表明,该方法能够实现高质量的故障自动分类识别。

Low resolution limits the quality and automation level of holospectrum technique in machinery diagnosis. To remedy this program, the basic definitions and methods of time series similarity matching are applied to the fault diagnosis, and combined nique. The holospectrum ries is used to express the with the idea of information fusion analysis for the rotating machinery of holospectrum techseries and its similarity measurement model are defined. Pseudo multidimensional time serotation panorama of the rotator system, and the holospectrum series similarity matching algorithm based on the searching strategy of the combination of the weaker triangle inequality and the B ＋ searching tree is used to achieve the fault diagnosis for the rotating machine. Experimental results show that the method proposed above can effectively achieve high quality automatic fault identification and classification.